Millions of Americans develop chest pain suggestive of coronary artery disease (CAD) each year and often receive non-invasive diagnostic testing such as myocardial perfusion scintigraphy (MPS) or, more recently, coronary computed tomographic angiography (CCTA). Such testing accounts for ~10% of the entire ionizing radiation burden to the U.S. population. The 200-site, 10,000 patient PROspective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) Trial is testing the hypothesis that an initial 'anatomic' (CCTA) testing strategy is superior in reducing major adverse cardiovascular events to usual care with an initial 'functional' stress testing strategy (using physician-selected stress imaging (MPS or echocardiography) or stress electrocardiography)) in low-intermediate CAD risk patients with chest pain. However, comprehensive comparison of testing strategies requires accurate assessment not just of cardiovascular events, but also weighting these against costs and risks for each strategy. The objective of this ancillary PROMISE Substudy to Assess For Effective dose of Radiation (PROMISE-SAFER) is to compare the radiation burden of initial anatomic vs. functional testing strategies for the evaluation of suspected CAD. Its primary aim is to determine whether cumulative radiation dose to symptomatic patients is decreased with an anatomic imaging strategy as compared to a functional testing strategy. While PROMISE will record basic measures of radiation burden from initial CCTA and MPS exams, the proposed study will develop a more accurate and comprehensive approach that is needed to accurately compare patients' cumulative radiation effective doses between testing strategies. This will include tracking subsequent tests for CAD evaluation and tests to evaluate incidental findings such as pulmonary nodules, and measured or imputed patient-specific dosimetry estimates from each follow-up test, with imputation performed using data from a large dose registry. Moreover, accurate estimation of radiation effective dose in CCTA will require determination of new conversion factors for contemporary CT scanners, to be performed using measurements obtained using solid-state radiation detectors in physical anthropomorphic phantoms. Additional aims are to compare projected cumulative lifetime attributable risk of cancer incidence, as determined using radioepidemiological models developed by the National Academies, between testing strategies, and to characterize the variability of radiation doses of cardiac imaging procedures, identifying patient, strategy, site, and regional factors predictive of high dose and repeated testing.